Follow-up Comment #9, sr#111026 (group administration):
[comment #8 comment #8:]
> [comment #7 comment #7:]
> > The AGPL itself does not require this kind of notice, as far as I can
tell,
>
> The AGPL requires that the program in question prominently offer every its
user the corresponding
Follow-up Comment #6, sr#111026 (group administration):
The spamming is not required by the AGPL, the text should simply be removed,
This has been reiterated so many times. I also think that someone else than
you gets assigned this task, since it is outright impossible to discuss
anything with
Follow-up Comment #2, sr#111026 (group administration):
Indeed, this was raised quite some time ago and still hasn't been fixed. The
notice is not required by the AGPL, and is just causing annoyance.
Bob suggested removing it, but that seems to have simmered into nothing.
Ineiev, no it would
Bandali reported that lists.gnu.org ran out of disk space, and
escalated that to sysadmin@.
There was a spike in spam on Monday as well .. so might be
afterquakes...
Thank you for this report. I don't know why but the permissions on
that one file were incorrect.
drwxrws--- 6 root gnueval 253 Nov 9 11:34 /srv/svn/gnueval/db
Yeah, I cannot explain it .. the gnueval repository is not heavily
used. I have one job that runs a `svn update' to send
E.g., https://svn.savannah.gnu.org/viewvc/?root=gnueval ->
An Exception Has Occurred
Python Traceback
Traceback (most recent call last):
File "/usr/lib/viewvc/lib/viewvc.py", line 4322, in main
request.run_viewvc()
File "/usr/lib/viewvc/lib/viewvc.py", line 268, in run_viewvc
Can we please turn this nuisance off? It is nothing related to the
AGPL in any shape or form.
You even get this utter garbage when doing an rsync?!
...
sv_membersh is part of Savane.
In order to download the corresponding source code of Savane, run
rsync -avz --cvs-exclude
> I believe users who authenticate do interact with sv_membersh
> in a way analogous to the frontend PHP code invoked through Apache.
>
> I don't agree that the AGPL requires this message to be shown on every
> cvs update. It's one possible interpretation, but not the only
Thank you for your help!
Happy hacking!
It looks like I need some coaching with CVS. It's been a long, long
time since I used it daily.
No worries, same here .. trying to recall.
> $ cvsu
> X manual/html_node
$ cvsu
cvsu: command not found
I assume you had a typo, but for what, I'm not sure.
>https://web.cvs.savannah.gnu.org/viewvc/groff/groff/manual/
If you open that URL you will find a directory:
https://web.cvs.savannah.gnu.org/viewvc/groff/groff/manual/html_node/
Yes, that directory exists.
alongside this one:
There is a stale directory in the GNU groff project's website.
https://www.gnu.org/software/groff/manual/html_node/
This corresponds to the groff 1.22.4 release, which is nearly 5 years
old.
The directory also shows up in ViewVC.
>> Savannah Hackers, would you please add Alfred?
>
>Alfred was added to gnustandards on 2023-06-18.
>
> Not as admin. Brandon is no longer active.
Done; I've also added you to the 'www' group where the generated output
is posted, in prep/.
Please be more
> Savannah Hackers, would you please add Alfred?
Alfred was added to gnustandards on 2023-06-18.
Not as admin. Brandon is no longer active.
> > And some rule on what can "automatically" go into the GCS without
your
> > explicit review and approval.
>
>Normally all changes in the GNU standards have to be approved by me,
>though it would be ok to fix a trivial typo.
>
> Fair enough. I'm still
[[[ To any NSA and FBI agents reading my email: please consider]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
> And some rule on what can "automatically" go into the GCS without
> *waves hand as volunteer* I could do it, it isn't that much work.
Thank you for volunteering -- please take up the responsibility.
Do you need anyone else's help to do that?
Only admin access to https://savannah.gnu.org/projects/gnustandards --
CCing savannah hackers.
And some rule
That sounds reasonable to me. I would assume that if webmasters don't
have access to change robots.txt, it would just be the fsf tech team.
They do. It is in www.
I'm still sorta opposing this .. we have always been transparant.
That search engines "have no business" is as saying that
Might be worth noting that www.gnu.org is mostly usable locally from
the CVS checkout as well, if one needs to look things over.
So a patch to www-discuss@ or whatever for a unpublished article would
be sufficient in postponing any publishing. It would just be a matter
of applying the patch, and
Le 10/05/2023 à 20:50, Alfred M. Szmidt a écrit :
> > You've not explained the actual problem. What are you trying to
> > solve?
>
> "it" is the www-commits list, which registers all changes to the www
> directory, including
> You've not explained the actual problem. What are you trying to
> solve?
"it" is the www-commits list, which registers all changes to the www
directory, including to pages that are not published yet. I suspect most
of the other *-commits lists deal with source code repositories,
Because it registers every single commit to www,
What is "it"? How is this different from _any_ -commits list we have?
including to working directories that webmasters have disallowed,
for instance */po/, /server/staging/, */workshop/, /prep/gnumaint/,
etc.
Ok, and?
Please
I was searching for an article with DuckDuckGo, and guess what
appeared on top of the results...
https://lists.gnu.org/archive/html/www-commits/2023-05/msg00082.html
and
https://lists.gnu.org/archive/html/www-commits/2023-05/msg00062.html
!!
What is the problem?
I understand
I've started using NonGNU Savannah recently. I connected to `
download.savannah.nongnu.org` just to explore it. I found that there's a
directory called `/srv/audio-video`, what's the purpose of that? Can
ordinary savannah users upload files to that directory?
>From the README:
This
Create a forth-years@, and then make a alias on Fencepost for 40years?
This is the "news" feature in savannah projects, which is
automatically fed into planet.gnu.org. But since Arnold doesn't
want to enable news, there has to be a way to "manually" submit a
news item for planet.
There is no such way, the front page of the planet is aggregated
feeds. So
A quick addendum, if gawk had used the Savannah news feature the news
entry would have shown up on planet.gnu.org -- planet.gnu.org already
tracks gawk news from Savannah.
Somehwere in the sidebar:
(feed) garpd @ Savannah
(feed) gawk @ Savannah
(feed) gcal @ Savannah
(feed) gcc @
>From planet.gnu.org:
'Please write to webmast...@gnu.org for feed aggregation requests or
suggestions.'
So please send this to webmasters@ and they will do the right thing,
but you will need to have a RSS feed of some sort.
[[[ To any NSA and FBI agents reading my email: please consider]]]
[[[ whether defending the US Constitution against all enemies, ]]]
[[[ foreign or domestic, requires you to follow Snowden's example. ]]]
What is the right URL to refer to the repo of the GNU C Intro and Ref
Lately, mail delivery by gnu.org mailing lists intermittently suffers
from very long delays. For example, currently emacs-devel and
help-gnu-emacs mail is delivered with delays between 45 minutes and
1.5 hours (based on admittedly small and potentially
non-representative sample).
Follow-up Comment #1, sr #110725 (project administration):
FWIW, You do not need a Savannah account to send a email to grub-devel@ --
they are to separate things.
As for the delay in your account submission, one of the Savannah admins will
need to look at that.
Sometimes, some of us subscribed here, can also ping the FSF admins in
other ways -- sysadmin@ is not a list that is easily subscribed to
AFAIK.
But the gripe about the FSF Status page is real: lately I find it less
and less useful in such cases. Which is too bad.
Just to give a small shout, I agree.
How about reusing the gnu-c-manual Savannah project; the old
gnu-c-manual git repository is then renamed to gnu-c-reference-manual
or old-gnu-c-manual. And put the new GNU C book into gnu-c-manual or
whatever title the book will have.
People are talking about using sourcehut to host Emacs development.
Can we run that on savannah, or some other VM on the same server?
On maybe a semi-related note, is't the FSF going to start some
sourceforge? Is that supposed to be sourcehut or something else? Is
there a point in setting
5) ... more?
I'd add that putthing all the project files somewhere accessibly, say
in old-gnu/, would be a useful. And move any gnu/ files associated
with the project to old-gnu/.
Sounds like a good idea overall.
I suggest iank@ or sysadmin@.
I think there is little point in continuing to discuss how you think
Savannah is run, it is up to the Savannah hackers how they decide what
they wish to work on, or not. You can call this "wrong" or a
"mistake" -- but it doesn't change the way how the GNU project works.
Moving forward, I
The GNU project is not some FSF "backyard", nor is it "controlled" by
the FSF. It is unfriendly to accuse either Ian, or the FSF in this
manner. We are not entield to the FSF's time or help, yet they help
us and take care of our equipment and go beyond that -- because our
goals are the same.
> This usually happens... instead of moving forward, we get a
> repertoire of defensive arguments that takes us nowhere in facing and
> tackling the real problems at GNU.org .
I can confirm that this is a recurring theme with regard to multiple Gnu
packages (not just savannah).
We, the GNU family, have a reason to celebrate today :)
I do not see it at that. It is a sad afair when one needs to
intervenenterveene in this manner.
Thank you also to Amin Bandali, Christopher Dimech and Francisco Vila.
Team work!
Ian was also instrumental in this.
I'm going to truncate my reply quite heavily, since many points are
either tangential to the problem at hand, or not pertinent to solving
it.
I could have created the repositories myself in 5 minutes, as I did
in OSDN, if I had access to that resource. We need to be in full
control of
Savannah has always been asking for help (about every year there is a
call for such), and has been in need of help -- hackers keep Savannah
going, but more hackers are needed to make it leap.
But for anything to happen, someone has to start doing something --
would you like to organize that work?
Sorry, I've waited too long. I have started the migration o the GNU
Health mercurial repository to OSDN. It's a pity, but it seems like the
requests over these years for the modernization of Savannah have not
been taken into consideration.
They have, but like all volunteer projects
>Then, for libre-sapienza (Savannah task #15792), the university
>and its students form a single entity: the administration
>of the university provides the students with some information
>for internal needs of the university.
>
> Are the students, or the
[1:text/plain Hide]
On Fri, Nov 20, 2020 at 05:03:05AM -0500, Alfred M. Szmidt wrote:
...
>> Suppose you write a free program that can run in a free GNU/Linux
>> distro and talks with Google Maps. With it, people can use Google
Maps
>> and n
> > The package must not refer the user to any nonfree software; in
> > other words, it must not say anything that in our judgment is likely
> > to lead or steer users towards running or installing nonfree
> > software on their machine.
>
> That is a good
> The package must not refer the user to any nonfree software; in
> other words, it must not say anything that in our judgment is likely
> to lead or steer users towards running or installing nonfree
> software on their machine. E.g., communicating with network
> services
> From reading the requirements, I think it is already clear that
> requirement is for software running on your machine, is it not?
It is not explicit about this. Some seem to have interpreted "not
refer the user to any nonfree software" as forbidding use with an
online
So I think the Savannah hosting rules need to be modified
to say that the criterion is about nonfree software to be run
on the user's own computers, and that it does not apply to
communication with services run by other people or entities.
Would someone like to work on a draft of
> How is that different from a latest version that is not documented,
> and free software? After all, you don't need to provide the source
> code for the free software server program.
Use the Gnu Affero GPL and the free software server program must be
provided for users of the
Hi Alfred
Right. Replace "ecosystem" with 'More and more of the software hosted at
GNU Savannah'.
It seems bad enough to me, in cases when the server's latest version is
documented by nobody and is proprietary. Then it is only its older revisions
that were successfully
> "A different case is when the program talks across a network with
> a server running on another machine, and the server is proprietary or
> has an unknown license; unless the two pieces of software make a
> single program (for example because they exchange complex data
>
This is I think better directed to the FSF sysadmins. CCing.
> Date: Tue, 28 Jul 2020 09:54:28 +0200 (CEST)
> From: Angelo Graziosi
>
> I noticed that reading the post online
(https://lists.gnu.org/archive/html/emacs-devel/2020-07/index.html, for
example), the email address in
Why not simply use a reference repository?
git clone --mirror git://git.gnu.org/emacs.git ~/emacs-reference
And when cloning:
git clone --reference ~/emacs-reference git://git.gnu.org/emacs.git
the ~/emacs-reference place can be accessible where ever you are using
it. This will work
That page doesn't say that savannah.gnu.org is only for GNU packages.
It seems to be written to cover savannah.nongnu.org. Many of the
sections are redundant and potentially misleading because GNU packages
are covered by stricter requirements.
I think the main issue is that
Here is content of that link.
Hosting requirements
Please read these usage terms carefully. If you don't follow them,
we will not accept your project; if we don't have enough
information determine whether your project follows these terms, we
will have to ask you for more
(Sorry for over-quoting; Zoltan's message didn't go to the list.)
It did? At least I got it ... maybe it was BCCed.
Zoltan, your current tone is not welcome here.
It is not a right to have an Savannah account, you have also not shown
anything that is suitably hosted on Savannah, nor is it
You can surley find the time to go through the files yourself to see
which ones need ammending.
You can surley find the time to go through the files yourself to see
which ones lack a copyright notice if you can find the time to write a
response here.
Isn't any mechanism of moderation?
This is the mechanism for moderation, Ineiev is doing exactly that.
I really feels coerced with all this, when I think I offer full
collaboration.
Savannah has a set of guidelines that are stricter than most other
hosting sites, if you cannot follow
[I sent this message a week ago but did not get a response.
The lack of any response is worrysome -- is anyone there?]
I tak zaczÄ
Å ÄwirzyÄ, ÄwikaÄ,
ÄwierkaÄ, czyrkaÄ, czykczyrikaÄ,
Że aż kogut na patyku
ZapiaÅ gniewnie: "Kukuryku!"
I sent a couple of volunteers your way,
> Is there any reason we can't shutdown the old frontend now? I can't
> think of anything that is still using it. I think the time has come
> that we can now shutdown halt the old frontend.
I moved the content of frontend:/root
into frontend0:/root/old-frontend-root -
it had
I have forwarded this to the Chapters/Planet admin(s).
The GNU GPL is the obvious choice, since that is what GNU plotutils
and Daniel Llorens changes are licensed under; it might be an wise
idea to also upgrade the license to GNU GPL version 3. It is also the
license that protects computer user freedom the best.
he copyright notice should list all
However my own opinion is that all of the lists may be subscribed
to by anyone. Many do. Many archive sites subscribe to the mailing
lists and archive the messages. Full archives of all of the
messages are therefore easily available elsewhere. The archive is
not of subscribers.
Follow-up Comment #2, sr #109076 (project administration):
FWIW, I forwarded this to Brandon Invergo ; who usually
takes care of issues like this these days.
But as Assaf notes, this should be brought up with gnu-advisory@, or with rms@
directly -- who is the person that
Comments, suggestions and improvements welcomed.
I like it. Maybe put it into the GNU maintainers guide as well?
Maybe incorperate it with the Free Software Directory somehow, there
could be an icon showing that a package is available in Guix; and
which version.
Itâs clear that CVS is a hindrance for such purposes so if thereâs
another possibility,
Ludo, why must we go through this yet again? Surely you know perfectly
well by now that there is no other possibility at present.
Not to mention that it isn't a problem with CVS;
It used to be the case I think, been long now, that you could just set
the CVSROOT envar to the new value, and run "cvs up -Ad" or similar to
update to a new CVS root, and keep the working copy. Otherwise,
replacing all CVS/Root files with the right thing is a simple hack ...
cvs -z3 -d:ext:am...@cvs.savannah.gnu.org:/web/www co www
For that to work, you need to be a member. Use anonymous checkout instead,
cvs -z3 -d:pserver:anonym...@cvs.savannah.gnu.org:/web/www co www
Or some such, see http://savannah.gnu.org/cvs/?group=www for details.
FYI, also down for inetutils.
Perhaps Savannah is not suited for the purpose of replacing Github,
but at the very least the FSF should sponsor sites like:
https://notabug.org/
which are based on free software and host exclusively free software
(unlike Github, which accepts proprietary programs).
It seems that
> It seems that http://notabug.org also accepts non-free programs, for
> example:
>
> https://notabug.org/nelis/hg-shopp-theme
>
> This has no license information what so ever. If this is in error or
> not, I don't know.
This should have not happened, according to
There is no license that limits commercial use and still compatible
with GPLv2 (and GNU savannah).
Or any free software license for that matter. All free software
licenses explicitly allow commercial use.
Nacho hasn't responded within a month, and our deadline is August
25th! After that date google code no longer works. Please help
us to properly set up the VM to run Allura!
I do not think that Nacho reads this list (and he wasn't included in
CC).
He was.
Nacho hasn't responded within a month, and our deadline is August
25th! After that date google code no longer works. Please help us to
properly set up the VM to run Allura!
I do not think that Nacho reads this list (and he wasn't included in
CC). CCing both Nacho, and Jose who might
However, we are not sure that Savannah would be ready to host the full
source code of the Replicant system, along with a few extra repositories
such as our website. In total, we expect that the full source code is
about 50 GiB and this will keep on increasing, but will certainly remain
Nacho: what about chapters? See below.
Not sure if there is more, but currently chapters has:
/dev/xvda2 99G 71G 23G 76% /
I'm getting the following error when trying to push, anyone know what
is wrong?
$ git push --verbose
Pushing to git://git.sv.gnu.org/womb/hacks.git
Counting objects: 3, done.
Delta compression using up to 4 threads.
Compressing objects: 100% (3/3), done.
Writing objects: 100% (3/3), 426 bytes | 0
Instead of getting emotional, have you tried doing `git commit
--amend'? That command is exactly created for situations like this!
My new project has been recently approved (gnu-social-mode), but due to
a new member's mistake there is a commit on the repository which is
malformed.
The proper way is to use `git commit -amend'; not rewrite history.
But what do you mean malformed? These are the few last commits to
Or you, if you're the only one working on that repository, you
could try to fix it using `git revert` (see the manual for that).
Yeah, I know about git-revert. But the wrong commit is so
small/simple that it doesn't take much to revert it.
Just use `git commit --amend' for that.
Follow-up Comment #3, sr #108178 (project administration):
I'm guessing that you might not have permissions to change the bug, or that
the specific project has set up some non-standard fields and names. Which bug
are you refering to by the way?
I have uploaded a project on 12th October in savannah-non-gnu for
review, and if it goes well, accept. But as I can see, its not yet
assigned to any one for review. My previous experience shows, this
process is generally much quicker. What can I do in this regard?
You should have
Do you have a record of past savannah account names and can tell
my the full names and email addresses of these guys?
arafune
kurt
nabe
Check the ChangeLog file, arafune is Ryuichi Arafune, kurt is probobly
Kurt Swanson, and nabe is Hidenobu Nabetani.
Sean Farley
Steven, I forwarded you (with you in CC) to sysadmin@, they should be
able to help you resolve this.
As for making things easier in the future, there is a wrapper script
to make uploading (and what not easier) called gnupload, available in
gnulib. You might want to check that out instead of hand
debug1: Next authentication method: publickey
debug1: Offering RSA public key: /home/erik/.ssh/id_rsa_savannah
Just a thought, it (kinda) looks like you are offering a private RSA
key and not the public one here. Are you entierly sure that
.ssh/id_rsa_savannah is the _public_ key and not
http://savannah.gnu.org/task/?7379
Please remove this page. I do not want to publish my information.
Sorry, but by posting to a public tracker you do agree to have such
information published. Same goes for this mailing list, which is a
public one with archives available to all
We've done. We saw : For manuals, we allow only GNU FDL. Does it
means, using savannah it will be forbid to license our works
(schematics, name, and any other descriptions of specifications)
under GNU GPL license ?
That should be fine, if all files have a proper copyright notice,
[1:text/plain Hide]
We know this question might be seemed stupid by many of us, but it is a
real question.
We are developping an open source project, certainty based on GNU-GPL
license, but applied to non-software.
Please don't say that the GNU GPL is a open source license, the
Follow-up Comment #5, sr #107999 (project administration):
Such a license is incompatible with the GPL, and free software in general.
Distributing it via Savannah is not personal usage.
___
Reply to this item at:
Follow-up Comment #1, sr #107999 (project administration):
I'm kinda curious why a dictionary takes up 11K per entry (I am assuming 90k
files equals one entry per file), not even GCIDE (a English dictionary) is
that big (about 60M uncompressed)... 1G should be able to encompass most
popular
Follow-up Comment #3, sr #107999 (project administration):
Then my immediate question is, what is the license of RAE? The bottom text
says:
Real Academia Española © Todos los derechos reservados
Which is a no-go for hosting on Savannah.
Hi!
You've posted to the wrong list, savannah-help-pub...@gnu.org is for
help regarding Savannah, which is a hosting platform for free
software.
Your problem would be better addressed if you posted your message to
the FreeRTOS, or LwIP help or bug lists.
Follow-up Comment #1, sr #107971 (project administration):
Savannah (including all other GNU and FSF machines) are being moved to
a new colo.
| The GNU/FSF servers are moving between February 22nd and March 1st,
| 2012. There will be service interruptions during that period.
| There will be a
Follow-up Comment #3, sr #107969 (project administration):
Savannah (including all other GNU and FSF machines) are being moved to
a new colo.
| The GNU/FSF servers are moving between February 22nd and March 1st,
| 2012. There will be service interruptions during that period.
| There will be a
I assume this is known, but the savannah web site isn't answering and
neither is the git repo (git.sv.gnu.org).
Savannah (including all other GNU and FSF machines) are being moved to
a new colo.
| The GNU/FSF servers are moving between February 22nd and March 1st,
| 2012. There will be
The question is not whether Android requires aditional software to be
useful but whether it can be used with only free software, for otherwise
Android-specific projects can't run on a fully free enviroment which is
a requirement in GNU Savannah.
A simple approach might be to do what
1 - 100 of 123 matches
Mail list logo